Reliability growth evaluation new method for Gompertz model based on Bayesian theory

The Bayesian method is adopted as an important and suitable approach to the small sample reliability analysis and evaluation, which can use all kinds of the system information. How to acquire proper prior distribution is a key problem in system's Bayesian reliability evaluation. The reliability growth Bayesian method for the Gompertz model was proposed, which only using rank matrix and quantile to can obtain the prior distribution based on the test information of reliability growth. Then combining with the small sample experiment data, the reliability parameters can be evaluated by Bayesian method. The simulation example shows that the method proposed is effective, and better than classical evaluation method. Starting from the results obtained for Gompertz model, this approach can be extended to more reliability growth models.